Commit Graph

228 Commits

Author SHA1 Message Date
hubertf 36ece38f58 Document that this is also the frontend for the Objective C compiler
(so "man -k objective" turns up *something*)
2001-01-09 00:25:05 +00:00
msaitoh 3fbb403b7d fix two bugs.
gcc original rev. 1.89:
	* combine.c (if_then_else_cond): If comparing against zero,
	just return thing being compared.

gcc original rev. 1.132:
	* combine.c (combine_simplify_rtx): Don't create an if_then_else
	unless both args are general_operand.
2000-12-27 13:32:09 +00:00
msaitoh 67a8732011 fix a bug from gcc original rev. 1.82
(jump_optimize_1): More accurately detect casesi insns.
2000-12-27 13:25:52 +00:00
msaitoh b2f875eeff apply some fixes from egcs-current
rtlanal.c	gcc original rev. 1.22
rtl.h		rev. 1.67
jump.c		rev. 1.47
	* rtlanal.c (regs_set_between_p): New function.
	* rtl.h (regs_set_between_p): Prototype it.
	* jump.c (jump_optimize): Use it instead of modified_between_p
	in the Sep 2 change.

fixes PR#11696
2000-12-27 13:23:59 +00:00
msaitoh 7b4b859811 apply gcc original rev. 1.116:
> combine.c (simplify_comparison): Fix typo.

LTU, LEU, GTU, "LEU" -> LTU, LEU, GTU, "GEU"
2000-12-15 23:42:22 +00:00
simonb 3ebe38e227 Use the ASM_OUTPUT_COMMON() and ASM_OUTPUT_LOCAL() macros from gcc
2.95.2's m68kelf.h.  Fixes problems with the size of common objects.

From Charles Hannum.
2000-12-14 06:38:50 +00:00
scw 4c12d23bdb Fix C++ static ctors/dtors by providing the same macros as defined
in i386/netbsd-elf.h.
2000-12-09 20:38:04 +00:00
scw 36f2e9920b Set the emulation target to m68kelfnbsd. 2000-12-07 22:46:04 +00:00
kleink 9cee15bd12 In the ELF toolchain, use __mcount instead of mcount. 2000-12-07 10:14:08 +00:00
thorpej a464410d30 If NBSD_EXEC_PREFIX is set, use it to set MD_EXEC_PREFIX rather
than going with just "/usr/libexec/".

XXX This wouldn't be so gross if the GCC people would give
up their #undef habit.
2000-11-09 06:27:07 +00:00
mycroft fa3aaf6d8b Modify LIB_SPEC and LIBGCC_SPEC so that -lc and -lgcc are not included when
-shared or -symbolic (as on other platforms).  Also, override LIBGCC_SPEC for
all platforms (where it would incorrectly include the .a file explicitly on
some).
This should fix Mozilla on macppc.
2000-10-23 13:19:35 +00:00
sommerfeld dc67ff641f When setting warn_format to a nonzero value, always use
warn_format = MAX(warn_format, N);
This prevents -Wnetbsd-format-audit -Wall from turning off stricter checks.
2000-10-18 13:57:49 +00:00
sommerfeld ac07e85117 Enhanced format checking, mk II:
New option -Wno-format-extra-args : suppress warning for:
	printf("foo", bar);
since that isn't interesting from the POV of this audit, and it's more
common than most in some code.

also, do format args checking for
	printf(Aflag ? "format 1: %s" : "format 2: %s", bar);
albeit in a half-hearted way for now (just check the "then" case)
2000-10-05 13:08:57 +00:00
sommerfeld ad8aa5a492 Enhanced format string checking.
If -Wformat is enabled, try harder to chase down compile-time constant
format strings.. capable of catching things like:
	const char foo[]="%s %d";
	printf(foo, 1, "2");

If the new option -Wnetbsd-format-audit (no, I don't like the name,
either) is enabled, complain about non-constant format strings we
can't verify.  In particular, for *v*printf-like functions (i.e.,
non-variadic functions which are passed a va_list), insist that the
format string come from a parameter declared as a format string
argument, so that functions which should be declared as printflike are
caught at compile time.
2000-10-04 19:21:28 +00:00
msaitoh 9308d053b8 gen_call() takes only two arguments (from egcs-current) 2000-09-13 19:51:22 +00:00
tsubai 84ecfb5e88 Call mcount for profiling. 2000-08-22 11:24:41 +00:00
msaitoh 559af6df3c fix two bugs:
- cc1 dumps core when "-pg" is used
	- gas produces a syntax error
2000-08-02 12:07:12 +00:00
mycroft a7bc813f46 Pull in a change critical to making the sparc64 compile:
Redefine ASM_OUTPUT_DWARF_ADDR_CONST() using ASM_OUTPUT_DWARF2_ADDR_CONST(), so
that the latter can take a string.
2000-07-24 02:30:40 +00:00
sommerfeld 85788f586b Pass -m21264 to assembler for -mcpu=21264a or -mcpu=ev67 2000-07-03 12:24:10 +00:00
matt ecd1681660 incease the biggest alignment to 8 bytes 2000-07-03 02:07:15 +00:00
matt 5335c7ffea fix serious recog bug. Basically addl2 was never used since the rtx_equal_p
was using the wrong operands.
2000-07-03 01:09:52 +00:00
bjh21 9fe4c9c75f Add arm26 support to GCC. This is only very slightly tested at present. 2000-07-02 12:31:21 +00:00
kristerw 4c7e15910e Disable "fancy math" the correct way (it's TARGET_DEFAULT that is used
for that kind of flags -- the TARGET_CPU_DEFAULT is used to choose the
cpu type to generate code for...)

Fixes PR 10357.
2000-06-19 19:48:37 +00:00
kristerw 49fd06e241 Backout the previous since it isn't necessary. This version of gcc
already uses longjmp for exceptions.
2000-06-09 19:11:43 +00:00
enami f838117776 Backout the previous since it isn't necessary. This version of gcc
already uses longjmp for exceptions for alpha due to the lack of
definition of INCOMING_RETURN_ADDR_RTX.
2000-06-09 08:16:01 +00:00
kristerw 3bf5c279b4 Use sjlj exceptions. 2000-06-08 20:23:47 +00:00
msaitoh d65ffb0be7 sync movsi_i* from egcs-current.
Now the cc1 doesn't say "fixed or forbidden register 18 (t) was
spilled for class T_REGS" anymore.
2000-05-22 08:28:04 +00:00
msaitoh e55cd9cb85 Sync *{mul,div}*i3 with egcs-current.
Now we can compile mount_mfs (newfs) with -O2!
2000-05-20 01:52:52 +00:00
is 8a79eef99f Add clobber of the condition code register. Patch by Richard Earnshaw.
This fixes PR 9613.
2000-05-16 22:04:12 +00:00
shin 9d63410108 use TARGET_DEFAULT to generate PIC code.
closes PR 8574.
2000-05-03 08:23:12 +00:00
shin 4c6474fd65 remove MIPS SOFTFLOAT (forgot to delete these lines). 2000-05-01 13:49:05 +00:00
thorpej 36e829de63 Don't enable CIX on ev6/21264, but add an ev67/21264a switch that does
enable CIX.  Taken roughly from egcs-current.
2000-04-03 16:47:05 +00:00
msaitoh 60ad3dbe9a set DWARF_LINE_MIN_INSTR_LENGTH to 2 because the length of instruction
is fixed to 16bits.
2000-03-27 18:40:41 +00:00
elric 17b91be06f Changed ev5 L1 cache latency from 2 to 3, as suggested by Bill Dorsey. 2000-03-16 22:46:10 +00:00
shin ac9896415f remove MIPS SOFTFLOAT. 2000-03-05 05:41:03 +00:00
tron c28b234bd7 Let "egcs" recognize strftime(3)'s "%F" conversion. Patch supplied by
NAKAJIMA Yoshihiro in PR bin/9487.
2000-02-27 22:41:22 +00:00
tsutsui 9e1d562a1f - tm.h is automatically generated by configure in gnu/dist/gcc,
so egcs/arch/mips/tm.h should not contain any #ifdefs
- Use -DTARGET_{BIG,LITTLE}_ENDIAN_DEFAULT (which are used in egcs-current)
  to switch mipseb/mipsel
2000-02-19 00:44:31 +00:00
tsutsui 5c8ea7f9f0 Add mipseb support to egcs distfiles. 2000-02-16 11:28:45 +00:00
tsutsui 9658fcaa61 Add support of mipseb target. 2000-02-16 11:23:48 +00:00
kristerw 46e954b5e5 C++ exceptions doesn't work in i386 shared libs when using sjlj exceptions
(as a.out does per default.)  This is because the compiler forgets to
restore the PIC register after a builtin_setjmp (i.e. when catching an
exception.)  This fix is taken from the gcc 2.95 branch.

Fixes PR#9110.
2000-02-01 19:29:33 +00:00
mycroft f6f389921a #define POSIX 2000-01-19 03:57:17 +00:00
msaitoh d64461f26b apply some patches from egcs-current (to make gcc more reliable) 2000-01-06 19:20:35 +00:00
msaitoh 6bba0f3f4a pass arg 4 of the qsort() correctly. 2000-01-02 11:01:02 +00:00
thorpej 4a9e81df67 Pass in the appropriate machine directive to the assembler if an -mcpu=...
option is given to the compiler.  It's silly to have to specify it twice
on a command line (e.g. -mcpu=ev56 -Wa,-mev56), especially considering
that if you don't, and the compiler emits e.g. a BWX instruction, the
assembler will treat it as a macro, open-coding an equivalent, which can
have some serious unwanted side-effects in some situations.

Also, don't treat e.g. 21164 and ev5 as equivalent; the assembler treats
them differently: 21164 enables PALcode-only instructions, while ev5
does not.
1999-12-03 06:37:47 +00:00
msaitoh f0af9ff8e6 add "system.h" 1999-11-26 07:55:14 +00:00
drochner 16dcb3a8fd more radical fix for the last problem, from the egcs/gcc mailing list
(http://egcs.cygnus.com/ml/gcc/1999-10/msg00400.html).
Log message:
          * combine.c (get_last_value): If the last set of a register
        is after subst_low_cuid, then we can not use it to determine
        the register's last value.
1999-10-25 19:30:19 +00:00
msaitoh 8b2c663a45 gcc for sh3 1999-10-07 17:27:35 +00:00
takemura cdabc2eb91 Hpcmips needs completely FPU instruction free user land.
XXX, this modification is kinda ad-hoc.
1999-09-29 13:19:13 +00:00
kleink 0b9965d236 For ELF, rename the profiling entry from mcount to __mcount to avoid clashing
with the user's namespace, also within file scope; for binary compatibility,
define a weak alias for the previous name to cover our tracks.
1999-09-27 09:47:44 +00:00
drochner cfab800438 Fix for a misoptimization, from the egcs/gcc mailing list
(http://egcs.cygnus.com/ml/gcc/1999-08/msg00760.html).
Log message:
        * combine.c (get_last_value): Don't look for earlier sets if the last
        known set is somewhere in between the insns being combined.
1999-09-04 11:18:40 +00:00